Build and Automate Pipelines CERTIFIED-DATA-SCIENCE-PRACTITIONER Practice Questions (Page 8)

Questions 36–39

1. 36expert · hard

   A large insurance company is struggling with its centralized data platform. The central data team is a bottleneck, and domain teams (claims, underwriting, finance) have deep knowledge of their data but no ownership. The company wants to adopt data mesh. However, the CTO is concerned about a lack of consistency and governance across domains. Which implementation strategy best addresses the CTO's concern while still adopting data mesh?

   • AAdopt a data fabric architecture instead, which provides centralized governance and unified access.
   • BImplement a federated governance model with global standards for interoperability, security, and data quality, while domain teams own their data products.
   • CHave each domain team create its own data lake with no shared standards.
   • DKeep the central team in charge of all data quality and security policies, while domain teams only build pipelines.
   Reveal Solution Discussion Select an answer first
2. 37foundation · easy

   In the context of a data warehouse, what does the 'T' in ETL stand for?

   • ATransfer
   • BTransformation
   • CTruncation
   • DTesting
   Reveal Solution Discussion Select an answer first
4. 38foundation · easy

   How do micro-services and APIs contribute to modular data pipeline components?

   • AThey require all components to be deployed as a single monolithic application
   • BThey allow each pipeline step to be developed, deployed, and scaled independently
   • CThey eliminate the need for data storage
   • DThey only work with on-premises infrastructure
   Reveal Solution Discussion Select an answer first
5. 39application · medium

   A healthcare organization is modernizing its data management architecture. It has data spread across on-premises databases, a cloud data warehouse, and SaaS applications. The goal is to provide analysts with a unified, real-time view of patient data without physically copying data from all sources into a single repository. Which architectural component is most appropriate for this requirement?

   • AA data virtualization layer that provides a unified query interface over the disparate sources.
   • BA set of point-to-point APIs between each source and the analyst's BI tool.
   • CA batch ETL pipeline that copies all data into the cloud data warehouse nightly.
   • DA data lake that ingests raw data from all sources for analysts to query.
   Reveal Solution Discussion Select an answer first